Matlab and Python wrap of conditional random field (CRF) and fully connected (dense) CRF for 2D and 3D image segmentation, according to the following papers:

[1] Yuri Boykov and Vladimir Kolmogorov, "An experimental comparison of min-cut/max-flow algorithms for energy minimization in vision", in IEEE TPAMI, 2004.

Pushmeet Kohli and Philip H.S. Torr. "Efficiently solving dynamic markov random fields using graph cuts", ICCV, 2005

[2] Philipp Krähenbühl and Vladlen Koltun, "Efficient inference in fully connected crfs with gaussian edge potentials", in NIPS, 2011.

[3] Kamnitsas et al. "Multi-scale 3D convolutional neural networks for lesion segmentation in brain MRI", in Proceeding of ISLES challenge, MICCAI, 2015.

This repository depends on the following packages:

• Maxflow http://www.cs.ucl.ac.uk/staff/V.Kolmogorov/software.html

• DenceCRF http://graphics.stanford.edu/projects/drf/

• 3D Fully Connected CRF https://github.com/Kamnitsask/dense3dCrf